Dissipative base connections for moment
frame structures in airports and
other transportation systems
Amit Kanvinde, UC Davis
Ahmad Hassan, UC Davis
17 January 2020
Project Team and Collaborators
UC Davis
- Amit Kanvinde, PI
- Ahmad Hassan, PhD Student
- Tomasz Falborski, former postdoc now faculty at Gdansk University, Poland
- Vince Pericoli, former postdoc now Engineer at Sandia Labs
PEER BIP Partner, Forell Elsesser
- Mason Walters
- Ali Roufegarinejad
- Geoff Bomba
UC Irvine (collaborator)
- Farzin Zareian
- Pablo Torres Rodas, now faculty at Univ San Francisco de Quito, Ecuador
Brigham Young University (adviser)
- Paul Richards
University College London (UCL) (collaborator)
- Carmine Galasso
- Biao Song, PhD Student
Acknowledgments
• PEER
• American Institute of Steel Construction
• Charles Pankow Foundation
• California Strong Motion Instrumentation
Program, CA Department of Conservation
Overview
• Steel Moment Resisting
Frames and buildings are
critical to airport (and other
transportation) infrastructure
• Research on column base
connections in SMRFs has
lagged research on other
connections
• Implications for connection as
well as frame design
Specific Issues
• Designing bases to be stronger
than columns is impractical and
expensive
• No information on systems with
weak bases
• No experimental data on several
common base connection details
• Design does not usually account
for interactions between base
connection and frame
Overall Research Plan
PEER (SIMULATION BASED SYSTEM STUDIES)
Component models for
strength/stiffness/hysteresis
Demonstrate frame performance with
dissipative/flexible bases
Methodology to design frame-base system with
such bases
Motivate research on ductile and repairable bases
AISC + CHARLES PANKOW
FOUNDATION
(EXPERIMENTAL COMPONENT
STUDIES)
• Untested details
• Unbonded dissipative
elements to localize
yielding
• Resilience
CA STRONG MOTION INSTRUMENTATION
PROGRAM (VALIDATION AND
BENCHMARKING)
• Moment Frame Buildings
• Range of foundation types
DEMANDS
COMPONENT
MODELS
MODEL VALIDATION
OUTCOMES
• Design methodology for Frames with Weak Bases
• Rigorous Consideration of base-frame interactions
• Details that make this possible + data on untested details
• Code changes (e.g., AISC 341)
• Design Guide One update, Design Manual Updates
Overall Research Plan
PEER (SIMULATION BASED SYSTEM STUDIES)
Component models for
strength/stiffness/hysteresis
Demonstrate frame performance with
dissipative/flexible bases
Methodology to design frame-base system with
such bases
Motivate research on ductile and repairable bases
AISC + CHARLES PANKOW
FOUNDATION
(EXPERIMENTAL COMPONENT
STUDIES)
• Untested details
• Unbonded dissipative
elements to localize
yielding
• Resilience
CA STRONG MOTION INSTRUMENTATION
PROGRAM (VALIDATION AND
BENCHMARKING)
• Moment Frame Buildings
• Range of foundation types
DEMANDS
COMPONENT
MODELS
MODEL VALIDATION
OUTCOMES
• Design methodology for Frames with Weak Bases
• Rigorous Consideration of base-frame interactions
• Details that make this possible + data on untested details
• Code changes (e.g., AISC 341)
• Design Guide One update, Design Manual Updates
Overall Research Plan
DEMANDS
COMPONENT
MODELS
MODEL VALIDATION
PEER (SIMULATION BASED SYSTEM STUDIES)
Component models for
strength/stiffness/hysteresis
Demonstrate frame performance with
dissipative/flexible bases
Methodology to design frame-base
system with such bases
Motivate research on ductile and
repairable bases
CA STRONG MOTION
INSTRUMENTATION PROGRAM
(VALIDATION AND BENCHMARKING)
• Moment Frame Buildings
• Range of foundation types
AISC + CHARLES PANKOW
FOUNDATION
(EXPERIMENTAL COMPONENT
STUDIES)
• Untested details
• Unbonded dissipative
elements to localize
yielding
• Resilience
OUTCOMES
• Design methodology for Frames with Weak Bases
• Rigorous Consideration of base-frame interactions
• Details that make this possible + data on untested details
• Code changes (e.g., AISC 341)
• Design Guide One update, Design Manual Updates
HILTI
(FEM SIMULATIONS)
• Biaxial Bending
UCL & UC DAVIS
(RELIABILITY STUDY)
• Exposed Base
Reliability
Analysis
Overall Research Plan
DEMANDS
COMPONENT
MODELS
MODEL VALIDATION
PEER (SIMULATION BASED SYSTEM STUDIES)
Component models for
strength/stiffness/hysteresis
Demonstrate frame performance with
dissipative/flexible bases
Methodology to design frame-base
system with such bases
Motivate research on ductile and
repairable bases
CA STRONG MOTION
INSTRUMENTATION PROGRAM
(VALIDATION AND BENCHMARKING)
• Moment Frame Buildings
• Range of foundation types
AISC + CHARLES PANKOW
FOUNDATION
(EXPERIMENTAL COMPONENT
STUDIES)
• Untested details
• Unbonded dissipative
elements to localize
yielding
• Resilience
OUTCOMES
• Design methodology for Frames with Weak Bases
• Rigorous Consideration of base-frame interactions
• Details that make this possible + data on untested details
• Strength Model for Biaxial Bending of Base Plates
• Calibration of Resistance Factors added in Design Equations
• Code changes (e.g., AISC 341)
• Design Guide One update, Design Manual Updates
HILTI
(FEM SIMULATIONS)
• Biaxial Bending
UCL & UC DAVIS
(RELIABILITY STUDY)
• Exposed Base
Reliability
Analysis
Overall Research Plan
DEMANDS
COMPONENT
MODELS
MODEL VALIDATION
PEER (SIMULATION BASED SYSTEM STUDIES)
Component models for
strength/stiffness/hysteresis
Demonstrate frame performance with
dissipative/flexible bases
Methodology to design frame-base
system with such bases
Motivate research on ductile and
repairable bases
CA STRONG MOTION
INSTRUMENTATION PROGRAM
(VALIDATION AND BENCHMARKING)
• Moment Frame Buildings
• Range of foundation types
AISC + CHARLES PANKOW
FOUNDATION
(EXPERIMENTAL COMPONENT
STUDIES)
• Untested details
• Unbonded dissipative
elements to localize
yielding
• Resilience
OUTCOMES
• Design methodology for Frames with Weak Bases
• Rigorous Consideration of base-frame interactions
• Details that make this possible + data on untested details
• Strength Model for Biaxial Bending of Base Plates
• Calibration of Resistance Factors added in Design Equations
• Code changes (e.g., AISC 341)
• Design Guide One update, Design Manual Updates
HILTI
(FEM SIMULATIONS)
• Biaxial Bending
UCL & UC DAVIS
(RELIABILITY STUDY)
• Exposed Base
Reliability
Analysis
PEER Objectives
• Development and calibration of component (hinge)
models for column base connections
• Nonlinear simulation of archetype frames with dissipative
bases
• Application of simulation results for design development
• Inform component experiments and interpretation
• Development of design examples for moment frames with
dissipative bases
PEER Objectives
• Development and calibration of component (hinge)
models for column base connections
• Nonlinear simulation of archetype frames with dissipative
bases
• Application of simulation results for design development
• Inform component experiments and interpretation
• Development of design examples for moment frames with
dissipative bases
Component Hinge Models (for exposed and
embedded type connections)
PHYSICS-BASED
FUNCTIONAL
FORM
CALIBRATION
RULES
CSMIP Project on Base Flexibility
PEER Objectives
• Development and calibration of component (hinge)
models for column base connections
• Nonlinear simulation of archetype frames with dissipative
bases
• Application of simulation results for design development
• Inform component experiments and interpretation
• Development of design examples for moment frames with
dissipative bases
FEMA P695 Parametric Study using base
connection models
W 24 62
W 24 94
W 24 94
W 24 131
W 24 131
W 24 162
W 24 131
W 24 162
W 24 131
W 24 207
W 24 207
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 14 426
W 24 162
W 24 207
W 24 146
W 24 176
W 24 131
W 24 162
W 24 131
W 24 131
W 24 84
W 24 94
W 24 62
W 24 103
W 24 103
W 21 57
W 21 57
W 21 73
W 21 73
W 21 68
W 24 84
W 27 94
W 27 94
W 27 94
W 30 116
W 30 116
W 30 108 W 30 124 W 33 169
W 33 169
W 33 169
W 33 169
W 33 169
W 33 169
W 33 169
W 33 169
W 33 141
W 33 141
W 33 141
W 33 141
W 33 141
W 33 141
W 30 108
W 30 108
W 30 108
W 30 108
W 24 62
W 24 62
W 30 132
W 30 132
W 30 132
W 30 116
W 30 116
W 30 116
W 30 116
W 27 94
W 27 94
W 24 84
W 24 84
concrete foundation
concrete foundation
8-story frame
Plan view
(all frames)
3 6.1m
30.5m
3 6.1m
42.7m
12-story frame
4-story frame
20-story frame
concrete foundation concrete foundation
panel zone
detail
panel zone
hysteretic springs
bilinear hysteretic
springs at RBS locations
elastic beam/column
elements
P-Delta columns
gravity loads
truss elements
bilinear hysteretic
springs at column ends
modified IMK deterioration base springs
with pinched hysteretic response
two springs
in series
BASE CONNECTIONS
MODELED BASED ON NEWLY
DEVELOPED APPROACHES
Key Results (Probabilities of Failure – P695)
Base Rotation Limit:
• Weak base design feasible
with W
0
=3
• These moments are up to
120% lower than 1.1RyMp
• For 8-20 story buildings,
rotation capacity of 0.05
needed (target for new
details)
• Fairly realistic to achieve
based on past data
Falborski et al., (2019 – in press) “The effect of base connection
strength and ductility on the seismic performance of steel
moment resisting frames,” Journal of Structural Engineering,
American Society of Civil Engineers.
PEER Objectives
• Development and calibration of component (hinge)
models for column base connections
• Nonlinear simulation of archetype frames with dissipative
bases
• Application of simulation results for design development
• Inform component experiments and interpretation
• Development of design examples for moment frames with
dissipative bases
Experimental Study (AISC/Pankow)
Phase I:
Base connections with Reliably Ductile Details as well
as Shallowly Embedded Details
• 7 tests
• Fall 2019 – Winter 2020
• Testing launches within next weeks
Phase II:
Untested details for Deeply Embedded Connections
• 7 tests
• Fall 2020
Design Guides and Wrapup
Fall 2021
Experimental Study (AISC/Pankow)
Phase I:
Base connections with Reliably Ductile Details as well
as Shallowly Embedded details
• 7 tests
• Fall 2019 – Winter 2020
• Testing launches within next weeks
x4 Tests
Reliably Ductile
Connection
Anchor Rods
Specifically
Detailed as
Below-Ground
Fuse
Reliably Ductile Connection (Mason, Ali, Geoff - BIP)
Reduced Diameter
Section
Upset Threaded Rod
Polyethylene Tape
-For Debonding-
Schematic Plan
Over-sized hole in
baseplate
Shear Lug
Rod isolated with tape
Baseplate and
threads remain
elastic
Ductile Behavior
M
θ
Elastic Loading
Yield Mechanism in Anchors
M
θ
Plastic
Elongation
(For clarity, shear key not shown)
Compression Yield Mechanism
M
θ
Elastic Unloading
M
θ
Ductile Behavior
(For clarity, shear key not shown)
Experimental Study (AISC/Pankow)
Phase I:
Base connections with Reliably Ductile details as well
as Shallowly Embedded Details
• 7 tests
• Fall 2019 – Winter 2020
• Testing launches within next weeks
Shallowly Embedded
Connection
x3 Tests
Overtopping Slab
Diamond
Blockout
Footing
Experimental Study (AISC/Pankow)
Phase II:
Untested details for Deeply Embedded Connections
• 7 tests
• Fall 2020
Embedded with
Welded
Reinforcement
Embedded With
Welded Shear Studs
Embedded With Top
Plate
Setup and Status
x7 Specimens Cast &
Cured
Setup in Process
Final Phase – Design Guide/Code Development:
- Design Guide One (~2021)
- Seismic Design Manual
- AISC 341
AISC/Pankow oversight committee
Final Phase – Design Guide/Code Development:
- Design Guide One (~2021)
- Seismic Design Manual
- AISC 341
AISC/Pankow oversight committee
Biaxial Bending (HILTI)
Problem:
• Interaction between 2 directions
(Bidirectional Effects of Seismic & Wind
Loads)
• Compromises strength of connection
• Bearing Stresses & Anchor Force
Distribution
• Not addressed by current guidelines
(AISC Design Guide 1)
Multiple
Anchors in
Tension
-Not Only 2-
Biaxial Bending (HILTI)
SOFTWARE VALIDATION
Validated against 2 series of existing experimental data
with various configurations (Uniaxial)
Accurate Predictions for Anchor Forces
BIAXIAL INSIGHTS
Bearing Stresses
Distribution of Axial
Forces
Fairly Representative
through Varying
• Geometry
• Moment Angles
• Level of Axial Load
• Connection Strength
Kanvinde et al.
(2014)
Gomez et al.
(2010)
Biaxial Bending (HILTI)
MODEL FORMULATION
Probing Bearing Stress distribution for error
reduction (including DG1 assumption)
Assume Linear Distribution of forces in anchors
Assume Neutral Axis Orientation
N.A
=
Loading
Rectangular Dist.
Triangular Dist.
T3
T2
T1
Linear
Distribution
Single Force
T1
T3
T2
T4
N.A
Bearing
Area
0.00
0.50
1.00
1.50
2.00
2.50
0.00 0.20 0.40 0.60 0.80 1.00 1.20
Hilti/Model
P/(1.7.fc.Ab)
Ratio vs Normalized Axial Force
T4
Ratio
Biaxial Bending (HILTI)
MODEL FORMULATION
Probing Bearing Stress distribution for error
reduction (including DG1 assumption)
Assume Linear Distribution of forces in anchors
Assume Neutral Axis Orientation
N.A
=
Loading
Rectangular Dist.
Triangular Dist.
T3
T2
T1
Linear
Distribution
Single Force
T1
T3
T2
T4
N.A
Bearing
Area
0.00
0.50
1.00
1.50
2.00
2.50
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00
Hilti/Matlab
Angle of Neutral Axis
Ratio vs Orientation Neutral Axis
T4
Ratio
Design Considerations for Exposed Column Base Connections
AISC Design Guide One
• FOUR expected failure modes:
Proposed Approach
• THREE Mechanics Based failure modes:
•
concrete
= 0.65 is used to determine the
imposed loads;
•
plate in bending
= 0.9 and
rod in tension
= 0.75
M
P
T
M
P
T
f
max
M
P
T
f
max
M
P
T
u,rod
f
max
•
concrete
= 1.0 is considered to determine the
imposed loads;
•
plate in bending
and
rod in tension
: to be
determined from Reliability Analysis
Reliability Study (UCL – UC Davis)
Reliability Analysis
• 59 representative design cases (P-M
pairs):
4 x 4-story frames (both exterior & interior
bases);
2 x design locations (LA & SAC);
3 x design load types (earthquake, wind &
gravity);
2 x cases from AISC DG1;
1 x case from SEAOC design manual.
• Sources of uncertainty:
Sectional geometries;
Material properties;
Applied loads;
Mechanical models.
• Limit-state functions:
Concrete bearing failure;
Base plate flexural yielding (compression or
tension side);
Anchor rods axial yielding.
• Monte-Carlo simulation:
Reliability index (β) as a function of
;
AISC DG1: β
plate
= 1.9
β
rods
= 2.2
concrete
= 1.0
concrete
= 0.65
DG1 design approach
Mechanics based
approach
BASE PLATE ANCHOR RODS
Mainly compression side
Mainly compression side
Only tension side
Thinner plate
Thicker plate
